summaryrefslogtreecommitdiffstats
path: root/conf
Commit message (Collapse)AuthorAgeFilesLines
* clang: Update to 16.0.1 releaseKhem Raj2023-03-291-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Brings following bugfixes * 42d1b276f779 Bump version to 16.0.1 * ebafcb86c35c [CMake] Respect variables for specifying host tools even without LLVM_USE_HOST_TOOLS set * 8de1b29840b5 [llvm-objdump] Fix help message for --print-imm-hex * 1ca4b5cfaa70 [sanitizer][win] Change cmdline check to allow double backslashs * 973cea755440 [BOLT] Search section based on relocation symbol * 140c68db4fe4 [libc++] Avoid ODR violations in __exception_guard * a18482ac5089 [BOLT][AArch64] Replace NOP with adrp in AdrRelaxationPass to preserve relative offsets. * c2ef3257df81 [BOLT][NFC] Remove C-style out of bounds array ref * 1c1ab11d6e1e [RISCV][MC] Adjust conditions to emit R_RISCV_ADD*/R_RISCV_SUB* pairs * 34194d85544f [test] Add some interesting cases to MC/RISCV/riscv64-64b-pcrel.s * ee1493c713d6 [test] Improve MC/RISCV/riscv64-64b-pcrel.s to demonstrate regression due to D132262 * 4f243f137859 [test] Improve MC/RISCV/riscv64-64b-pcrel.s * 1f9ea2d3f045 [X86] AMD Genoa (znver4) Scheduler model update * 0b4106274631 [BOLT] Reject symbols pointing to section end * 14ed120b2307 No longer issue static lambda pedantic warning for pre-c++2b compat * 32b8cc7031f0 [BOLT] Fix data reoder for aarch64 * 867c59c2da17 [Coroutines] Pass size parameter for deallocation function when qualified * 6dc69d034845 [libc++][ranges] Fix incorrect integer typedef in `elements_view` test. * 99af55f91f5d [libc++][format] Fix a missing include in tests. * 8a80823218a8 [libunwind][PowerPC] Fix saving/restoring VSX registers on LE systems * c5b23ab2a31a JITLink: Add missing EHFrame NULL terminator on aarch64/ELF * 2352a1b1e917 Reapply [lld][flang] Add exceptions for Flang runtime libraries on MinGW. * 37ea87b88d07 No longer issue pedantic warning about pre-c++2b compat * a70565f7de2e [libunwind][Modules] Add unwind_arm_ehabi.h and unwind_itanium.h to the unwind module) * 56edf062bdb6 [dsymutil] dsymutil produces broken lines info (probably) with LTO on mac * bec9a607690b [compiler-rt] Add missing #else clause to fix the build on NetBSD. * 3c0850337bd8 [X86] Fix encoding for ATOMIC_LOGIC_OP * ba22382d514b [X86] Add negative test for D145930 * 2861fa247395 [Driver][FreeBSD] Simplify ARM handling * 1172ed57d823 [analyzer] Fix crashing getSValFromInitListExpr for nested initlists * 830229ee918d [WebAssembly] Add auto-upgrade for renamed intrinsics * b5f9ea64b207 [X86] Support load/store for bf16 in avx * bb6a97c15aa2 [AArch64] fix bug #55005 handle DW_CFA_GNU_NegateRAState * aef3508f89c4 [flang] Fixed uninitialized std::unique_ptr dereference. * fbf716ffe163 [flang] Fixed restrictions checking for OpenACC loop-associated constructs. * 2ecbe73ceeda [flang] Fix dereference of std::optional with no value * 0e227de08e32 [LLDB][ObjectFileELF] Correct the return type of Reloc{Offset,Addend}32 * 7d28293f17fa [LLDB][ObjectFileELF] Correct the return type of RelocOffset64 and RelocAddend64 * 6d3b7679ef21 [LLDB][ObjectFileELF] Support LoongArch64 in ApplyReloctions * a824efcd0158 [BPF] Improve pruning to avoid generate more types in BTF * ab86d147e852 [libunwind][AArch64] Unbreak building with GNU assembler * 526102b37e59 [InstCombine] Canonicalize icmp eq pow2 more thoroughly * 7049d5895700 [InstCombine] Add additional test for icmp eq/ne with bool load (NFC) * b3ea3484c063 [Pipelines] Restore old DAE position in LTO pipeline * 72cb90bd7b62 [PhaseOrdering] Add test for DAE/GlobalDCE interaction (NFC) Signed-off-by: Khem Raj <raj.khem@gmail.com>
* layer.conf: Add qt5 layer to dynamic layer listKhem Raj2023-03-221-0/+2
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* Revert "webkitgtk: Use gcc for now"Khem Raj2023-03-211-3/+0
| | | | This reverts commit 20a8cff8ba0ce02773946cd7ba398fc08969ce1e.
* frr: Use gcc when compiling for rv32/rv64Khem Raj2023-03-201-0/+3
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* nodejs: Disable enum-constexpr-conversion warning as errorKhem Raj2023-03-201-0/+5
| | | | | | | Clang-16 adds this new warnings which triggers on nodejs, disable it for now Signed-off-by: Khem Raj <raj.khem@gmail.com>
* opengl-es-cts,vulkan-cts,pidgin-sipe: Disable new warnings as errorsKhem Raj2023-03-201-0/+10
| | | | | | | Clang-16 finds more warnings which are treated as errors, lets disable them for now Signed-off-by: Khem Raj <raj.khem@gmail.com>
* clang: Update to 16.0.0Khem Raj2023-03-201-1/+1
| | | | | | | | Changes are described here [1] [1] https://discourse.llvm.org/t/llvm-16-0-0-release/69326 Signed-off-by: Khem Raj <raj.khem@gmail.com>
* android-tools: remove recipe and dynamic-layerEtienne Cordonnier2023-03-151-1/+0
| | | | | | | The recipe has been patched to remove the clang dependency and moved to meta-openembedded at https://github.com/openembedded/meta-openembedded/tree/master/meta-oe/dynamic-layers/selinux/recipes-devtool/android-tools Signed-off-by: Etienne Cordonnier <ecordonnier@snap.com>
* nonclangable.conf: Add new exceptions for alternative Linux kernel recipesJon Mason2023-02-221-0/+2
| | | | | | | Other Linux kernel recipes are seeing the same objcopy issue. Add entries for them to work around the issue here as well. Signed-off-by: Jon Mason <jdmason@kudzu.us>
* linux-fslc-imx: Use gnu objcopy with clangKhem Raj2023-02-151-0/+1
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* pixman: Use gcc on mipsKhem Raj2023-01-291-0/+2
| | | | | | | libatomic is needed to link after -lomp when -fopenmp is used, it needs clang driver level change Signed-off-by: Khem Raj <raj.khem@gmail.com>
* clang: Update to 15.0.7Khem Raj2023-01-161-1/+1
| | | | | | | | | | | | * 8dfdcc7b7bf6 [libc++] Fix memory leaks when throwing inside std::vector constructors * 939f5a337113 libc++: bring back the unsigned in the return type in wcstoull_l * 948cadd6d424 [RegAllocFast] Handle new debug values for spills * 1095870e8ced [wasm-ld] Define a `__heap_end` symbol marking the end of allocated memory. * 67fd0d2af4bf [TypePromotion] Add truncate in ConvertTruncs when the original truncate type is not extend type * 74d3ba1af5c0 [X86] Don't zero out %eax if both %al and %ah are used * a8af9f679231 Bump version to 15.0.7 Signed-off-by: Khem Raj <raj.khem@gmail.com>
* u-boot-coral: Use gcc toolchainKhem Raj2023-01-091-0/+1
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* linux-fslc-lts: Use binutils objcopyKhem Raj2023-01-091-0/+1
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* mybw: Use gnu objcopy alwaysKhem Raj2023-01-081-6/+4
| | | | | | | Its failing on most of architectures only known working arch seems to be arm Signed-off-by: Khem Raj <raj.khem@gmail.com>
* sysbench: Add -no-integrated-as to cflags with clangKhem Raj2023-01-081-0/+1
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* nonclangable.conf: Use binutils objcopy for BSP kernelsKhem Raj2023-01-041-0/+11
| | | | | | llvm objcopy does not yet work fully for all kernels Signed-off-by: Khem Raj <raj.khem@gmail.com>
* u-boot: Use gcc for variants of u-boot from BSP layersKhem Raj2023-01-041-0/+21
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* pixman: Use gcc on aarch64Khem Raj2023-01-041-1/+3
| | | | | | | Some upgrades have regressed pixman with clang on aarch64, until this is sorted, lets use gcc for it as a workaround Signed-off-by: Khem Raj <raj.khem@gmail.com>
* systemd: Add -Wl,--no-warn-rwx-segment with clangKhem Raj2023-01-031-0/+1
| | | | | | Remove the Os workaround, this is no longer needed with systemd 252+ Signed-off-by: Khem Raj <raj.khem@gmail.com>
* ruby,cairo: Add -Wl,--no-warn-rwx-segment with clangKhem Raj2023-01-031-0/+3
| | | | | | | This helps with configure tests which treat this warning as error otherwise Signed-off-by: Khem Raj <raj.khem@gmail.com>
* mybw: Use gnu objcopy on rv32Khem Raj2023-01-031-0/+1
| | | | | | | | Fixes /usr/bin/mybw', 'TOPDIR/build/tmp/work/riscv32-yoe-linux/mybw/0.0+gitAUTOINC+f4bdeee126-r0/package/usr/bin/.debug/mybw']' returned non-zero exit status 1. Subprocess output:riscv32-yoe-linux-llvm-objcopy: error: Link field value 37 in section .rela.dyn is not a symbol table Signed-off-by: Khem Raj <raj.khem@gmail.com>
* layer.conf: update LAYERSERIES_COMPAT for mickledoreMartin Jansa2023-01-031-1/+1
| | | | | | | * oe-core switched to mickedore in: https://git.openembedded.org/openembedded-core/commit/?id=57239d66b933c4313cf331d35d13ec2d0661c38f Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
* nonclangable.conf: Add new exceptionsAleksey Smirnov2022-12-121-1/+10
| | | | | | | | | Add "erlang" for all architectures (found on riscv64 and x86-64) Add "grub" for x86-64 architecture Use proper OBJCOPY for "linux-yocto" for all architectures Signed-off-by: Aleksey Smirnov <aleksey.smirnov@yadro.com> Signed-off-by: Khem Raj <raj.khem@gmail.com>
* u-boot: Always use gcc compiler to buildKhem Raj2022-12-051-0/+3
| | | | | | Clang-15 fails to compile u-boot on mips right now. Signed-off-by: Khem Raj <raj.khem@gmail.com>
* linux-yocto: Use gnu objcopy on mipsKhem Raj2022-12-051-0/+3
| | | | | | | Fixes Subprocess output:mips-yoe-linux-llvm-objcopy: error: Link field value 22 in section .rel.dyn is not a symbol table Signed-off-by: Khem Raj <raj.khem@gmail.com>
* clang: Update to 15.0.6Khem Raj2022-12-051-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | Drop backported patches already available in this release This brings following changes * 088f33605d8a [CodeGen][ARM] Fix ARMABIInfo::EmitVAAarg crash with empty record type variadic arg * abcd0341d846 [ELF] Handle GCC collect2 -plugin-opt= on Windows * e6e61e9b2ef7 Revert "Reapply: Add an error message to the default SIGPIPE handler" * 25a36ca5c791 Bump version to 15.0.6 * 154e88af7ec9 Bump version to 15.0.5 * a39989663758 [VectorUtils] Skip interleave members with diff type and alloca sizes. * dc8f6ffc3bf2 [lldb] Get rid of __STDC_LIMIT_MACROS and __STDC_CONSTANT_MACROS * 392963bb1daf [lldb] Fix 'error: non-const lvalue...' caused by SWIG 4.1.0 * 68799e789fc5 [GlobalOpt] Don't remove inalloca from varargs functions * 11c3a21f8d1b [analyzer] Workaround crash on encountering Class non-type template parameters * 0988addf2680 Link libclangBasic against libatomic when necessary. * 4c3d83810ad7 Link liblldCOFF against libatomic when necessary * d75ae21044ad Set LLVM_ATOMIC_LIB variable for convenient linking against libatomic * 6750e341b076 [TypePromotion] Replace Zext to Truncate for the case src bitwidth is larger * 58ba50a52ede [cmake] Add missing CMakePushCheckState include to FindLibEdit.cmake * 931b6d51d84e Reenable POSIX builtin library functions in gnu2x mode * c8e7a87b1ed6 [CMake] Fix -Wstrict-prototypes Signed-off-by: Khem Raj <raj.khem@gmail.com>
* rsync: Use gcc for nowKhem Raj2022-11-271-0/+3
| | | | | | | | Needed until [1] is merged into oe-core [1] https://patchwork.yoctoproject.org/project/oe-core/patch/20221123215941.2411460-1-raj.khem@gmail.com/ Signed-off-by: Khem Raj <raj.khem@gmail.com>
* pixman: Do not use clang assembler for nowKhem Raj2022-11-251-0/+1
| | | | | | | | | | | | | Results in assembly file errors e.g. | <instantiation>:1:1: error: unknown directive | .func fname | ^ | <instantiation>:2:1: note: while in macro instantiation | pixman_asm_function fname | ^ Signed-off-by: Khem Raj <raj.khem@gmail.com>
* clang: Upgrade to 15.0.4Khem Raj2022-11-041-1/+1
| | | | | | | | | | | | | | | | | | | | | | Brings these changes * 5c68a1cb1231 AMDGPU: Make various vector undefs legal * 80a9fc840b1b [clang][Sema] Fix a clang crash with btf_type_tag * 08bd84e8a635 [SPARC] Make calls to function with big return values work * 9d46557baa84 Take memset_inline into account in analyzeLoadFromClobberingMemInst * dd711a939122 [lld-macho] Canonicalize personality pointers in EH frames * 3010b7e00000 [clang][driver] Remove dynamic gcc-toolset/devtoolset logic * db68723804fd [clangd] Return earlier when snippet is empty * 2d5c43ad4844 [lldb] Automatically unwrap parameter packs in template argument accessors * 1e1c5204c259 [SystemZ] Relase notes for LLVM 15 * dccd0613025a [ELF] Suppress "duplicate symbol" when resolving STB_WEAK and STB_GNU_UNIQUE in different COMDATs * 5834fe66318b [AutoUpgrade] Fix remangling when upgrading struct return type * 5388da13992c Bump version to 15.0.4 * 455e1d765ad6 [InstCombine] Bail out of casting calls when a conversion from/to byval is involved. * ceee53ce564c [SROA] Don't speculate phis with different load user types * 086365be7ce0 [test] Make Linux/sem_init_glibc.cpp robust Signed-off-by: Khem Raj <raj.khem@gmail.com>
* mybw: USe gnu objcopy on riscv64Khem Raj2022-10-271-0/+1
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* mybw: Use libgcc to link on x86_64Khem Raj2022-10-261-0/+2
| | | | | | | | Fixes x86_64-yoe-linux-ld: /usr/src/debug/glibc/2.36-r0/stdio-common/printf_fphex.c:198: undefined reference to `__letf2' since compiler-rt does not provide the 'tf' functions Signed-off-by: Khem Raj <raj.khem@gmail.com>
* mybw: Use objcopy from binutils on mipsKhem Raj2022-10-221-0/+4
| | | | | | | | | Use libgcc on x86 instead of compiler-rt llvm objcopy errors out so disable it for now Subprocess output:mips-yoe-linux-llvm-objcopy: error: Link field value 42 in section .rel.dyn is not a symbol table Signed-off-by: Khem Raj <raj.khem@gmail.com>
* clang: Upgrade to 15.0.3Khem Raj2022-10-191-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | Brings following changes * 4a2c05b05ed0 [ARM] Fix for MVE i128 vector icmp costs. * f6af95770615 [clangd] Block clang-tidy misc-const-correctness check * 687250913265 [libc++] Fix std::function's handling of blocks under Objc ARC * b4840279846e [CMake] Provide Findzstd module * 6fba7854a2f0 [llvm] [test] Add missing canonicalization of LLVM_ENABLE_ZSTD * e3b0721b5b73 [llvm] [lit] Fix use_lld() to respect llvm_shlib_dir * dfef316bb3d6 [clang-tidy] Fix a false positive in readability-simplify-boolean-expr * d35bc70e8251 [AArch64][SVE] Fix AArch64_SVE_VectorCall calling convention * f3c5289e7846 Revert "Recommit "[SCEV] Look through single value PHIs." (take 3)" * 6599b6e5ae9e Fix LLDB build on old Linux kernels (pre-4.1) * 02129eab7d58 [Syntax] avoid using c++17 features on 15.x branch * fc47af8c9140 Bump version to 15.0.3 * 27e075fcfad1 [Syntax] Fix macro-arg handling in TokenBuffer::spelledForExpanded * 359ef0c93240 [clangd] Improve inlay hints of things expanded from macros * bd5722b87b5a [X86] Remove AVX512VP2INTERSECT from Sapphire Rapids. * 73ea64f30411 [clangd] Avoid scanning up to end of file on each comment! * 61fa70903191 [LoopVersioning] Invalidate SCEV for phi if new values are added. * c0748fe6dec5 [clang-tools-extra] [clangd] Respect llvm_shlib_dir in tests Signed-off-by: Khem Raj <raj.khem@gmail.com>
* fmt: Use libgcc on x86_64 as wellKhem Raj2022-10-121-0/+1
| | | | | | | | compiler-rt does not provide the TF functions for x86 or x86_64 since it does not support TF functions on architectures which does not support 128-bit long double. Signed-off-by: Khem Raj <raj.khem@gmail.com>
* fmt: Use libgcc on x86Khem Raj2022-10-031-0/+1
| | | | | | | This is because compiler-rt does not provide int128 builtins which fmt uses Signed-off-by: Khem Raj <raj.khem@gmail.com>
* layers: Add langdale to LAYERSERIES_COMPATKhem Raj2022-09-291-1/+1
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* clang: Bump to 15.0.2Khem Raj2022-09-281-1/+1
| | | | | | | | | | | | | | | | | | | | | | | Brings these changes * 451e3b68306d Bump version to 15.0.2 * 9d3635c8675f Exclude check-polly-unittests and check-polly-isl from check-all * a26ef3c19ede [Hexagon] Add defaulted operator= to classes with defaulted copy ctor * f1ad3abf75f0 [MachineCycle][NFC] add a cache for block and its top level cycle * 8d802f78fa8c [libc++] Keep unary_function and binary_function in C++17 for one more release * 5d9fa4d16e0e SPIRV: Fix compilation in NDEBUG. * b73d2c8c720a [LV] Keep track of cost-based ScalarAfterVec in VPWidenPointerInd. * 38b5fa73d585 [LV] Move new epilog-vectorization-widen-inductions.ll to AArch64 dir. * c079a29305bb [LV] Add tests for epilogue vectorization with widened inductions. * d64394b81d6e [libc++] Always query the compiler to find whether a type is always lockfree * c6d2e8bcc033 [docs] Fix build-docs.sh * 43b5b04f27f1 [libcxx] Bump libc++ version to 15.0.1 * d4bada99c069 [CodeGen] Don't zero callee-save registers with zero-call-used-regs (PR57692) * 433f2aaa32bd [lit] Set shlibpath_var on OpenBSD * 063556d02def [clang(d)] Include/Exclude CLDXC options properly * 046d5b917bcd [Libomptarget] Revert changes to AMDGPU plugin destructors Signed-off-by: Khem Raj <raj.khem@gmail.com>
* gnutls,harfbuzz: No need for adding -latomic on rv64Khem Raj2022-09-181-2/+0
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* Revert "concurrencykit: Do not need to use -no-integrated-as anymore on arm"Khem Raj2022-09-181-0/+8
| | | | This reverts commit 320e91209c7000729dfafbc84a72dbd755f18483.
* lttng-ust: Use clang to compile it again on rv64Khem Raj2022-09-181-2/+0
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* mpich: Drop using libgcc unwinder on x86Khem Raj2022-09-181-1/+0
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* libbsd: Add more info about the induced errors due to libbsdKhem Raj2022-09-171-0/+2
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* kexec-tools: Do not need -no-integrated-as anymore on x86Khem Raj2022-09-171-8/+0
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* aufs-util,libc-bench: Drop forcing to use libgcc undwinder on armKhem Raj2022-09-171-2/+0
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* concurrencykit: Do not need to use -no-integrated-as anymore on armKhem Raj2022-09-171-8/+0
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* libsdl2: Drop using -no-integrated-as on armKhem Raj2022-09-171-6/+0
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* libvorbis,mpg123: Drop using -fno-fast-math with clangKhem Raj2022-09-171-4/+0
| | | | Signed-off-by: Khem Raj <raj.khem@gmail.com>
* qemu,libffi,openssl: Drop clang kludges for armKhem Raj2022-09-171-17/+0
| | | | | | They are not needed anymore Signed-off-by: Khem Raj <raj.khem@gmail.com>
* openssl: Drop using -no-integrated-asKhem Raj2022-09-171-5/+0
| | | | | | clang-15 is fine without it Signed-off-by: Khem Raj <raj.khem@gmail.com>